# Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts

arXiv: https://arxiv.org/abs/2104.00887 · HTML: https://mlsanity.com/paper/2104.00887/

## Key facts

- Cited by 46 papers in the map (data through 14 Oct 2026).
- Builds on 37 of its 57 references that are in the map.
- First posted on arXiv 2 Apr 2021.

## Builds on (most cited first, 30 of 37)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features](https://mlsanity.com/paper/1905.04899/) (2019)
- [Large Scale GAN Training for High Fidelity Natural Image Synthesis](https://mlsanity.com/paper/1809.11096/) (2018)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [Learning Deep Features for Discriminative Localization](https://mlsanity.com/paper/1512.04150/) (2015)
- [Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization](https://mlsanity.com/paper/1703.06868/) (2017)
- [Similarity of Neural Network Representations Revisited](https://mlsanity.com/paper/1905.00414/) (2019)
- [Self-training with Noisy Student improves ImageNet classification](https://mlsanity.com/paper/1911.04252/) (2019)
- [StarGAN v2: Diverse Image Synthesis for Multiple Domains](https://mlsanity.com/paper/1912.01865/) (2019)
- [StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation](https://mlsanity.com/paper/1711.09020/) (2017)
- [Self-Attention Generative Adversarial Networks](https://mlsanity.com/paper/1805.08318/) (2018)
- [Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks](https://mlsanity.com/paper/1703.10593/) (2017)
- [Universal Style Transfer via Feature Transforms](https://mlsanity.com/paper/1705.08086/) (2017)
- [GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond](https://mlsanity.com/paper/1904.11492/) (2019)
- [Few-Shot Unsupervised Image-to-Image Translation](https://mlsanity.com/paper/1905.01723/) (2019)
- [Learning De-biased Representations with Biased Representations](https://mlsanity.com/paper/1910.02806/) (2019)
- [Deep Photo Style Transfer](https://mlsanity.com/paper/1703.07511/) (2017)
- [Photorealistic Style Transfer via Wavelet Transforms](https://mlsanity.com/paper/1903.09760/) (2019)
- [A Closed-form Solution to Photorealistic Image Stylization](https://mlsanity.com/paper/1802.06474/) (2018)
- [Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels](https://mlsanity.com/paper/2101.05022/) (2021)
- [Multi-Content GAN for Few-Shot Font Style Transfer](https://mlsanity.com/paper/1712.00516/) (2017)
- [NSML: Meet the MLaaS platform with a real-world case study](https://mlsanity.com/paper/1810.09957/) (2018)
- [Separating Style and Content for Generalized Style Transfer](https://mlsanity.com/paper/1711.06454/) (2017)
- [A Unified Feature Disentangler for Multi-Domain Image Translation and Manipulation](https://mlsanity.com/paper/1809.01361/) (2018)
- [Artistic Glyph Image Synthesis via One-Stage Few-Shot Learning](https://mlsanity.com/paper/1910.04987/) (2019)

## Built on by (most cited first, 30 of 46)

- [TextDiffuser: Diffusion Models as Text Painters](https://mlsanity.com/paper/2305.10855/) (2023)
- [AnyText2: Visual Text Generation and Editing With Customizable Attributes](https://mlsanity.com/paper/2411.15245/) (2024)
- [FontDiffuser: One-Shot Font Generation via Denoising Diffusion with Multi-Scale Content Aggregation and Style Contrastive Learning](https://mlsanity.com/paper/2312.12142/) (2023)
- [Improving Diffusion Models for Scene Text Editing with Dual Encoders](https://mlsanity.com/paper/2304.05568/) (2023)
- [Diff-Font: Diffusion Model for Robust One-Shot Font Generation](https://mlsanity.com/paper/2212.05895/) (2022)
- [Few-Shot Font Generation by Learning Fine-Grained Local Styles](https://mlsanity.com/paper/2205.09965/) (2022)
- [Look Closer to Supervise Better: One-Shot Font Generation via Component-Based Discriminator](https://mlsanity.com/paper/2205.00146/) (2022)
- [CF-Font: Content Fusion for Few-shot Font Generation](https://mlsanity.com/paper/2303.14017/) (2023)
- [VecFusion: Vector Font Generation with Diffusion](https://mlsanity.com/paper/2312.10540/) (2023)
- [XMP-Font: Self-Supervised Cross-Modality Pre-training for Few-Shot Font Generation](https://mlsanity.com/paper/2204.05084/) (2022)
- [DreamText: High Fidelity Scene Text Synthesis](https://mlsanity.com/paper/2405.14701/) (2024)
- [WeditGAN: Few-Shot Image Generation via Latent Space Relocation](https://mlsanity.com/paper/2305.06671/) (2023)
- [Few shot font generation via transferring similarity guided global style and quantization local style](https://mlsanity.com/paper/2309.00827/) (2023)
- [Few-shot Font Generation with Weakly Supervised Localized Representations](https://mlsanity.com/paper/2112.11895/) (2021)
- [SVG Vector Font Generation for Chinese Characters with Transformer](https://mlsanity.com/paper/2206.10329/) (2022)
- [FontTransformer: Few-shot High-resolution Chinese Glyph Image Synthesis via Stacked Transformers](https://mlsanity.com/paper/2210.06301/) (2022)
- [FASTER: A Font-Agnostic Scene Text Editing and Rendering Framework](https://mlsanity.com/paper/2308.02905/) (2023)
- [VQ-Font: Few-Shot Font Generation with Structure-Aware Enhancement and Quantization](https://mlsanity.com/paper/2308.14018/) (2023)
- [HFH-Font: Few-shot Chinese Font Synthesis with Higher Quality, Faster Speed, and Higher Resolution](https://mlsanity.com/paper/2410.06488/) (2024)
- [Few-shot Font Generation by Learning Style Difference and Similarity](https://mlsanity.com/paper/2301.10008/) (2023)
- [DeepCalliFont: Few-shot Chinese Calligraphy Font Synthesis by Integrating Dual-modality Generative Models](https://mlsanity.com/paper/2312.10314/) (2023)
- [APRNet: Attention-based Pixel-wise Rendering Network for Photo-Realistic Text Image Generation](https://mlsanity.com/paper/2203.07705/) (2022)
- [CalliffusionV2: Personalized Natural Calligraphy Generation with Flexible Multi-modal Control](https://mlsanity.com/paper/2410.03787/) (2024)
- [MegaHan97K: A Large-Scale Dataset for Mega-Category Chinese Character Recognition with over 97K Categories](https://mlsanity.com/paper/2506.04807/) (2025)
- [DGFont++: Robust Deformable Generative Networks for Unsupervised Font Generation](https://mlsanity.com/paper/2212.14742/) (2022)
- [MX-Font++: Mixture of Heterogeneous Aggregation Experts for Few-shot Font Generation](https://mlsanity.com/paper/2503.02799/) (2025)
- [DA-Font: Few-Shot Font Generation via Dual-Attention Hybrid Integration](https://mlsanity.com/paper/2509.16632/) (2025)
- [Beyond Patches: Global-aware Autoregressive Model for Multimodal Few-Shot Font Generation](https://mlsanity.com/paper/2601.01593/) (2026)
- [FontNet: Closing the gap to font designer performance in font synthesis](https://mlsanity.com/paper/2205.06512/) (2022)
- [SGCE-Font: Skeleton Guided Channel Expansion for Chinese Font Generation](https://mlsanity.com/paper/2211.14475/) (2022)

## Abstract (arXiv)

A few-shot font generation (FFG) method has to satisfy two objectives: the generated images should preserve the underlying global structure of the target character and present the diverse local reference style. Existing FFG methods aim to disentangle content and style either by extracting a universal representation style or extracting multiple component-wise style representations. However, previous methods either fail to capture diverse local styles or cannot be generalized to a character with unseen components, e.g., unseen language systems. To mitigate the issues, we propose a novel FFG method, named Multiple Localized Experts Few-shot Font Generation Network (MX-Font). MX-Font extracts multiple style features not explicitly conditioned on component labels, but automatically by multiple experts to represent different local concepts, e.g., left-side sub-glyph. Owing to the multiple experts, MX-Font can capture diverse local concepts and show the generalizability to unseen languages. During training, we utilize component labels as weak supervision to guide each expert to be specialized for different local concepts. We formulate the component assign problem to each expert as the graph matching problem, and solve it by the Hungarian algorithm. We also employ the independence loss and the content-style adversarial loss to impose the content-style disentanglement. In our experiments, MX-Font outperforms previous state-of-the-art FFG methods in the Chinese generation and cross-lingual, e.g., Chinese to Korean, generation. Source code is available at https://github.com/clovaai/mxfont.